docs: update chlog after 2.0.8 and 2.0.9 (#18569)

Author: Borda

src/lightning/app/CHANGELOG.md

@@ -34,6 +34,21 @@ The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/).
 ### Fixed
 
 
+## [2.0.9] - 2023-09-14
+
+### Fixed
+
+- Replace LightningClient with import from lightning_cloud ([#18544](https://github.com/Lightning-AI/lightning/pull/18544))
+
+
+## [2.0.8] - 2023-08-29
+
+## Changed
+
+- Change top folder ([#18212](https://github.com/Lightning-AI/lightning/pull/18212))
+- Remove `_handle_is_headless` calls in app run loop ([#18362](https://github.com/Lightning-AI/lightning/pull/18362))
+
+
 ## [2.0.7] - 2023-08-14
 
 ### Changed

src/lightning/fabric/CHANGELOG.md

@@ -129,9 +129,6 @@ The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/).
 - Allow using iterable-style datasets with TPUs ([#17331](https://github.com/Lightning-AI/lightning/pull/17331))
 
 
-- On XLA, avoid setting the global rank before processes have been launched as this will initialize the PJRT computation client in the main process ([#16966](https://github.com/Lightning-AI/lightning/pull/16966))
-
-
 - Increased the minimum XLA requirement to 1.13 ([#17368](https://github.com/Lightning-AI/lightning/pull/17368))
 
 
@@ -177,18 +174,10 @@ The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/).
 
 ### Fixed
 
-- Fixed FSDP full-precision `param_dtype` training (`16-mixed`, `bf16-mixed` and `32-true` configurations) to avoid FSDP assertion errors with PyTorch < 2.0 ([#18278](https://github.com/Lightning-AI/lightning/pull/18278))
-
 
 - Fixed issue where running on TPUs would select the wrong device index ([#17227](https://github.com/Lightning-AI/lightning/pull/17227))
 
 
-- Fixed issue where Fabric would not initialize the global rank, world size, and rank-zero-only rank after initialization and before launch ([#16966](https://github.com/Lightning-AI/lightning/pull/16966))
-
-
-- Removed false positive warning when using `fabric.no_backward_sync` with XLA strategies ([#17761](https://github.com/Lightning-AI/lightning/pull/17761))
-
-
 - Removed the need to call `.launch()` when using the DP-strategy (`strategy="dp"`) ([#17931](https://github.com/Lightning-AI/lightning/pull/17931))
 
 
@@ -198,13 +187,29 @@ The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/).
 - Fixed FSDP re-wrapping the module root when the user had manually wrapped the model ([#18054](https://github.com/Lightning-AI/lightning/pull/18054))
 
 
-- Fixed model parameters getting shared between processes when running with `strategy="ddp_spawn"` and `accelerator="cpu"`; this has a necessary memory impact, as parameters are replicated for each process now ([#18238](https://github.com/Lightning-AI/lightning/pull/18238))
+- Fixed issue where unexpected exceptions would leave the default torch dtype modified when using true precision settings ([#18500](https://github.com/Lightning-AI/lightning/pull/18500))
+
 
+## [2.0.9] - 2023-09-14
+
+### Fixed
 
 - Fixed an issue causing the `_FabricOptimizer.state` to remain outdated after loading with `load_state_dict` ([#18488](https://github.com/Lightning-AI/lightning/pull/18488))
 
 
-- Fixed issue where unexpected exceptions would leave the default torch dtype modified when using true precision settings ([#18500](https://github.com/Lightning-AI/lightning/pull/18500))
+## [2.0.8] - 2023-08-29
+
+### Changed
+
+- On XLA, avoid setting the global rank before processes have been launched as this will initialize the PJRT computation client in the main process ([#16966](https://github.com/Lightning-AI/lightning/pull/16966))
+
+### Fixed
+
+- Fixed model parameters getting shared between processes when running with `strategy="ddp_spawn"` and `accelerator="cpu"`; this has a necessary memory impact, as parameters are replicated for each process now ([#18238](https://github.com/Lightning-AI/lightning/pull/18238))
+- Removed false positive warning when using `fabric.no_backward_sync` with XLA strategies ([#17761](https://github.com/Lightning-AI/lightning/pull/17761))
+- Fixed issue where Fabric would not initialize the global rank, world size, and rank-zero-only rank after initialization and before launch ([#16966](https://github.com/Lightning-AI/lightning/pull/16966))
+- Fixed FSDP full-precision `param_dtype` training (`16-mixed`, `bf16-mixed` and `32-true` configurations) to avoid FSDP assertion errors with PyTorch < 2.0 ([#18278](https://github.com/Lightning-AI/lightning/pull/18278))
+
 
 ## [2.0.7] - 2023-08-14
 

src/lightning/pytorch/CHANGELOG.md

@@ -135,9 +135,6 @@ The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/).
 - Allow using iterable-style datasets with TPUs ([#17331](https://github.com/Lightning-AI/lightning/pull/17331))
 
 
-- On XLA, avoid setting the global rank before processes have been launched as this will initialize the PJRT computation client in the main process ([#16966](https://github.com/Lightning-AI/lightning/pull/16966))
-
-
 - Increased the minimum XLA requirement to 1.13 ([#17368](https://github.com/Lightning-AI/lightning/pull/17368))
 
 
@@ -221,9 +218,6 @@ The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/).
 
 ### Fixed
 
-- Fixed FSDP full-precision `param_dtype` training (`16-mixed`, `bf16-mixed` and `32-true` configurations) to avoid FSDP assertion errors with PyTorch < 2.0 ([#18278](https://github.com/Lightning-AI/lightning/pull/18278))
-
-
 - Fixed an issue with reusing the same model across multiple trainer stages when using the `DeepSpeedStrategy` ([#17531](https://github.com/Lightning-AI/lightning/pull/17531))
 
 
@@ -233,34 +227,35 @@ The format is based on [Keep a Changelog](http://keepachangelog.com/en/1.0.0/).
 - Fixed FSDP re-applying activation checkpointing when the user had manually applied it already ([#18006](https://github.com/Lightning-AI/lightning/pull/18006))
 
 
-- Fixed an issue that prevented the use of custom logger classes without an `experiment` property defined ([#18093](https://github.com/Lightning-AI/lightning/pull/18093))
-
-
-- Fixed model parameters getting shared between processes when running with `strategy="ddp_spawn"` and `accelerator="cpu"`; this has a necessary memory impact, as parameters are replicated for each process now ([#18238](https://github.com/Lightning-AI/lightning/pull/18238))
-
-
-- Fixed setting the tracking uri in `MLFlowLogger` for logging artifacts to the MLFlow server ([#18395](https://github.com/Lightning-AI/lightning/pull/18395))
+- Fixed issue where unexpected exceptions would leave the default torch dtype modified when using true precision settings ([#18500](https://github.com/Lightning-AI/lightning/pull/18500))
 
 
-- Fixed redundant `iter()` call to dataloader when checking dataloading configuration ([#18415](https://github.com/Lightning-AI/lightning/pull/18415))
+## [2.0.9] - 2023-09-14
 
+### Fixed
 
 - Fixed an issue that wouldn't prevent the user to set the `log_model` parameter in `WandbLogger` via the LightningCLI ([#18458](https://github.com/Lightning-AI/lightning/pull/18458))
-
-
 - Fixed the display of `v_num` in the progress bar when running with `Trainer(fast_dev_run=True)` ([#18491](https://github.com/Lightning-AI/lightning/pull/18491))
+- Fixed `UnboundLocalError` when running with `python -O` ([#18496](https://github.com/Lightning-AI/lightning/pull/18496))
+- Fixed visual glitch with the TQDM progress bar leaving the validation bar incomplete before switching back to the training display ([#18503](https://github.com/Lightning-AI/lightning/pull/18503))
+- Fixed false positive warning about logging interval when running with `Trainer(fast_dev_run=True)` ([#18550](https://github.com/Lightning-AI/lightning/pull/18550))
 
 
-- Fixed issue where unexpected exceptions would leave the default torch dtype modified when using true precision settings ([#18500](https://github.com/Lightning-AI/lightning/pull/18500))
-
-
-- Fixed `UnboundLocalError` when running with `python -O` ([#18496](https://github.com/Lightning-AI/lightning/pull/18496))
+## [2.0.8] - 2023-08-29
 
+### Changed
 
-- Fixed visual glitch with the TQDM progress bar leaving the validation bar incomplete before switching back to the training display ([#18503](https://github.com/Lightning-AI/lightning/pull/18503))
+- On XLA, avoid setting the global rank before processes have been launched as this will initialize the PJRT computation client in the main process ([#16966](https://github.com/Lightning-AI/lightning/pull/16966))
+- Fix inefficiency in rich progress bar ([#18369](https://github.com/Lightning-AI/lightning/pull/18369))
 
+### Fixed
 
-- Fixed false positive warning about logging interval when running with `Trainer(fast_dev_run=True)` ([#18550](https://github.com/Lightning-AI/lightning/pull/18550))
+- Fixed FSDP full-precision `param_dtype` training (`16-mixed` and `bf16-mixed` configurations) to avoid FSDP assertion errors with PyTorch < 2.0 ([#18278](https://github.com/Lightning-AI/lightning/pull/18278))
+- Fixed an issue that prevented the use of custom logger classes without an `experiment` property defined ([#18093](https://github.com/Lightning-AI/lightning/pull/18093))
+- Fixed setting the tracking uri in `MLFlowLogger` for logging artifacts to the MLFlow server ([#18395](https://github.com/Lightning-AI/lightning/pull/18395))
+- Fixed redundant `iter()` call to dataloader when checking dataloading configuration ([#18415](https://github.com/Lightning-AI/lightning/pull/18415))
+- Fixed model parameters getting shared between processes when running with `strategy="ddp_spawn"` and `accelerator="cpu"`; this has a necessary memory impact, as parameters are replicated for each process now ([#18238](https://github.com/Lightning-AI/lightning/pull/18238))
+- Properly manage `fetcher.done` with `dataloader_iter` ([#18376](https://github.com/Lightning-AI/lightning/pull/18376))
 
 
 ## [2.0.7] - 2023-08-14